Public health surveillance systems in Ethiopia have undergone significant structural changes, yet their operational reliability remains methodologically under-evaluated. A robust, quantitative framework for forecasting system performance is absent, limiting proactive interventions. This study aimed to develop and validate a time-series forecasting model to assess the reliability of the national surveillance system and to project future performance under current operational conditions. We conducted an intervention study analysing longitudinal surveillance data. A Seasonal Autoregressive Integrated Moving Average (SARIMA) model, specified as $\text{SARIMA}(p,d,q)(P,D,Q)_s$, was fitted to historical completeness and timeliness metrics. Model parameters were estimated using maximum likelihood, and forecasts were generated with 95% prediction intervals. The fitted SARIMA(1,1,1)(0,1,1)_12 model indicated a significant negative trend in system timeliness, with a forecasted decline in on-time reporting of 15.2 percentage points over the projection period. Forecast uncertainty, represented by prediction interval width, increased substantially beyond the immediate forecast horizon. The surveillance system exhibits a statistically significant decline in reliability, which is projected to continue without intervention. The forecasting model provides a novel tool for pre-emptive system assessment. Implement the forecasting methodology for routine monitoring and allocate resources to districts identified as high-risk for reporting failures. Future work should integrate environmental and agricultural covariates to enhance model specificity. surveillance, forecasting, reliability, SARIMA, public health, Ethiopia This paper introduces a novel application of SARIMA modelling for the predictive reliability assessment of public health surveillance, providing a quantitative tool for pre-emptive system strengthening.